CropCop is a closed‑set plant‑health recognition system covering 120 operational classes, built from a rigorously audited dataset of 109,107 images after removing 3,233 duplicate relationships. The model, based on a fine‑tuned DINOv3 ConvNeXt‑Tiny, achieves 98.51% accuracy and 96.87% macro‑F1 on a locked internal test, while a quantised MobileNetV4 variant reaches 98.46% accuracy and 96.23% macro‑F1 in a 22.60 MiB runtime artifact. Validation‑only post‑training quantisation and a compact ExecuTorch/XNNPACK PTE ensure high fidelity between the trained model and its deployed form, with minimal decision changes between the INT8 graph and the final artifact.
By Rana Muhammad Ahmed, Sabahat Abbas
PairCLIP‑SWA is a single‑backbone deepfake detector that fine‑tunes an OpenCLIP ViT‑L/14 visual tower using a class‑balanced corpus created by CLIP‑similarity‑guided subsampling and then applies stochastic weight averaging over the fine‑tuning trajectory. Trained on the IEEE SP Cup 2025 (DF‑Wild‑Cup) benchmark, the weight‑averaged model achieves 97.04 % accuracy, 98.64 % AU‑ROC and a 4.07 % equal error rate—outperforming the best published three‑backbone ensemble. The study shows that averaging stabilizes validation performance, prevents misranking on disjoint test sets, and that CLIP pre‑training, rather than ensembling, is the key driver of accuracy.
By Md. Nakib, Sudipto Sarkar, A. J. A. Jubayer Talukder, Mohammad Ariful Haque
The paper evaluates event-triggered recording strategies for static surveillance cameras, comparing six motion-based triggers, a CNN detector, strided detection, and hybrid approaches on a benchmark of seven scenes. Detector-based triggers reduce idle frame storage to 19‑28% versus 36‑89% for motion triggers, and remain robust under camera jitter, keeping under 29% idle frames. In addition, the study shows that increasing YOLO11n input resolution improves detection accuracy more than larger models, and that longer track memory boosts tracking IDF1, though detection remains the limiting factor.
By Shahran Rahman Alve
arXiv:2610. 39676v1 Announce Type: cross Abstract: Event cameras offer microsecond temporal resolution, high dynamic range and no motion blur, but referring video object segmentation (RVOS) lacks event-stream datasets and models.
By Sergei Fedorchenko, Max Zehnder, Nikola Zubi\'c, Rong Zou, Davide Scaramuzza
arXiv:2609.05762v1 Announce Type: cross
Abstract: UAV image collections contain spatially and temporally related frames, yet semantic-segmentation benchmarks commonly split them at image level. Such...
By Tarek Rahman, Nazim-E-Alam, Md Kishor Morol, Jannatun Noor
The paper explores whether rapid, salience-driven behavior can be detected in high‑frequency Counter‑Strike 2 telemetry by reconstructing players’ perceptual states with a millisecond‑precision visibility engine. Using a 60–120 ms reaction window, the authors derive fight‑or‑flight labels and train two telemetry‑native models—EMBER (gradient boosting) and ORBIT (custom transformer)—which outperform a majority‑class baseline and a large language model while requiring far less inference time and cost. Adding a short post‑trigger window further improves accuracy to 85.15%, suggesting a potential probe of the shift from rapid to deliberative control, though the study remains hypothesis‑generating rather than mechanistic.
By Emanuel Dopater, Daniel Gavlas, Luk\'a\v{s} Jochymek, Kannan Paul, Miloslav Szczypka, David Voj\'a\v{c}ek, Josef \v{Z}\'a\v{c}ek, Jan Plucar, Radek Svoboda, Ivan Zelinka
arXiv:2610.26933v1 Announce Type: new
Abstract: An event camera emits events at a rate given by the scene motion, object texture, illumination, its own contrast thresholds, and other settings. Across...
By Kavyansh Tyagi, Andreas Ziegler
General-purpose object detectors lose accuracy on UAV footage, where targets span only a handful of pixels and onboard compute is limited. Prior work composes independently-validated architectural tec...
arXiv:2607. 20129v1 Announce Type: new Abstract: Quantized small autoregressive reasoning models can enter long, repetitive, or unproductive trajectories, yet inference-time compute is usually allocated without observing how a trajectory develops.
By El Hassane Ettifouri, Ayoub Belfatmi, Mahaman Sanoussi Yahaya Alassan, Walid Dahhane
arXiv:2609.14560v1 Announce Type: new
Abstract: General-purpose object detectors lose accuracy on UAV footage, where targets span only a handful of pixels and onboard compute is limited. Prior work c...
By Quratulain Nayeem, Fahmina Taranum, Mohammed Mudassir Uddin
ObserverBench is a benchmark framework that evaluates whether internal mechanistic estimators—called observers—are suitable for guiding interventions, control, or safety actions in language models. It separates estimation accuracy from the loss incurred by the chosen action, showing that accurate predictions do not always lead to better decisions. Experiments on GPT‑2‑small, Qwen2.5‑7B, Gemma‑2‑9B‑it, and Qwen3.5‑9B demonstrate that observers trained on action loss can reduce deployment loss, while traditional metrics like AUROC may rank monitors differently from actual performance.
By Vijay Erramilli
The study investigates whether the number of discrete class‑separability jumps (phase transitions) observed during ResNet fine‑tuning can predict final test accuracy. Across 75 experiments on four benchmarks (CIFAR‑10, CIFAR‑100, TinyImageNet, CIFAR‑10‑C) and three ResNet variants, a strong negative correlation is found on standard i.i.d. datasets (r = −0.84 on CIFAR‑10, r = −0.87 on CIFAR‑100), while the correlation weakens under distributional stress. Additional analyses show that the transition count retains predictive power after controlling for architecture depth and outperforms other training‑curve signals on in‑distribution benchmarks, though it is dominated by other signals on stressed datasets.
By Arunan J